Disclosure of potential conflicts of interest is found at the end of this article.
Author contributions: X.Y.: conception and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing; H.W.: collection and/or assembly of data, data analysis and interpretation; X.Z.: collection and/or assembly of data; S.Z.: collection and/or assembly of data; Q.Z.: conception and design, financial support; S.D.: conception and design, financial support, manuscript writing, final approval of manuscript. X.Y. and H.W. contributed equally to this work.
First published online in STEM CELLSEXPRESS January 7, 2011.
It has been established that exogenous expression of four transcription factors (Oct4, Klf4, Sox2, and c-Myc) can reprogram mammalian somatic cells to pluripotent states. Further studies demonstrated that such induced pluripotent stem cells (iPSCs) could be generated with fewer exogenous transcription factors, facilitated by endogenous expression of reprogramming factors and/or synthetic small molecules. Here, we reported identification of a new small molecule, a protein arginine methyltransferase inhibitor AMI-5, which enabled Oct4-induced reprogramming of mouse embryonic fibroblasts in combination with transforming growth factor (TGF)-β inhibitor A-83-01. The Oct4-induced iPSCs were shown similar to mouse embryonic stem cells with respect to typical pluripotency criteria. More importantly, they were shown to give rise to liveborn pups through tetraploid complementation assays, demonstrating the high quality of full reprogramming induced by this condition. Furthermore, this study suggests that regulation of protein arginine methylation might be involved in the reprogramming process. STEM CELLS 2011;29:549–553
Ectopic expression of four transcription factors (Oct4, Sox2, Klf4, c-Myc, and/or Nanog, Lin28) can reset the epigenome of somatic cells with restricted differentiation potentials to pluripotent states [1–8]. Especially enabled with small molecules that can enhance reprogramming and/or functionally substitute reprogramming factors, the direct reprogramming can be achieved with two or three transcription factors, and even just Oct4 from neural progenitor cells (NPCs) and fibroblasts [9–18]. Those small molecules include direct epigenetic modifiers, such as histone methyltransferase G9a inhibitor, DNA methyltransferase inhibitors, histone demethylase inhibitor, and histone deacetylase inhibitors, as well as signaling pathway modulators, such as mitogen-activated protein kinase inhibitor, glycogen synthase kinase 3β inhibitors, and transforming growth factor-β (TGF-β) pathway inhibitors [10–13, 16, 19–23]. Their use and subsequent characterizations not only facilitate generation of induced pluripotent stem cells (iPSCs) but also provide insights into the mechanisms of reprogramming. Toward the ultimate goal of defining a cocktail of small molecules for generation of iPSCs, identifying new and more effective reprogramming enhancing small molecules would be highly valuable.
Here, we report identification of a new small molecule, a protein arginine methyltransferase (PRMT) inhibitor AMI-5 , which enabled Oct4-induced reprogramming of mouse embryonic fibroblasts (MEFs) in combination with TGF-β inhibitor A-83-01, demonstrating AMI-5's superiority. It suggests that regulation of protein arginine methylation might be involved in the reprogramming process. Furthermore, this simple condition may serve as a powerful platform to identify additional new compounds and study the mechanism of Oct4-induced reprogramming. Moreover, the Oct4-induced mouse iPSCs (miPSCs-O) were shown to give rise to liveborn pups through tetraploid complementation assays, demonstrating the high quality of full reprogramming induced by this condition.
MATERIALS AND METHODS
Primary Oct4-EGFP (enhanced green fluorescent protein) MEFs were infected with Oct4-expressing retrovirus (Addgene, Cambridge, MA) and treated with 1 μM A-83-01 (Stemgent, San Diego, CA) plus other compounds till GFP+ (green fluorescent protein) colonies appeared. Manually picked iPSC clones were expanded and characterized through immunocytochemistry, reverse transcription polymerase chain reaction (RT-PCR), and bisulfite methylation assay (Zymo Research Corp., Orange, CA). Their pluripotency was investigated by in vitro differentiation, chimera construction, and tetraploid complementation. (See Supporting Information for details)
Primary MEFs containing a transgenic Oct4-EGFP reporter were transduced with retrovirus carrying Oct4. A small chemical library (∼100 small molecules, a collection of epigenetic modifiers and signaling pathway modulators) was screened in the presence of TGF-β receptor inhibitor A-83-01. Neither the Oct4 transduction alone nor the addition of A-83-01 generated any iPSC in our screening condition, providing a clean background. Remarkably, we observed 1-4 GFP+ iPSC colonies in the wells treated with AMI-5 (5 μM; Fig. 1A) between the 30th and 40th days (Fig. 1B). The reprogramming efficiency was estimated at ∼0.02%, which is comparable with Oct4-induced reprogramming of NPC (0.01%) . Seven GFP+ miPSC-O colonies were picked and successfully expanded in conventional mouse embryonic stem cell (mESC) culture condition (Fig. 1C), and two of them were further characterized.
The stably expanded miPSC-O cells were morphologically indistinguishable to mESCs, stained positive for alkaline phosphatase (ALP) (Fig. 1D) and expressed typical pluripotency markers such as Nanog, Sox2, and stage-specific embryonic antigen 1 (SSEA1) proteins (Fig. 1E). The endogenous Oct4 expression was verified by RT-PCR (Fig. 2A). We detected no exogenous viral Oct4 expression in the miPSC-O (Fig. 2B), suggesting retroviral silencing (another characteristic of full reprogramming). Besides Nanog, Klf4, and Sox2, many undifferentiated ES cell marker genes were expressed in the miPSCs-O at levels equivalent to those of the R1 mESCs, including Esg1, Ecat1, Eras, Gdf3, Fgf4, Rex1, Utf1, and Cripto (Fig. 2A). Using PCR primers specific for the retroviral vectors, we confirmed the integration of the viral Oct4 transgene in the miPSCs-O genome but no contaminations from the c-Myc, Klf4, or Sox2 transgenes (Fig. 2C).
Bisulfite genomic sequencing analyses revealed that Nanog and Oct4 promoter regions were highly demethylated in the miPSC-O, similar to those in mESCs, whereas they were hypermethylated in MEFs (Fig. 2D). These results confirmed the epigenetic remodeling that happened during reprogramming and the reactivation of pluripotency in the miPSC-O.
The differentiation potential of the miPSC-O was investigated through embryoid body differentiation. Immunocytochemistry demonstrated that they could effectively differentiate into ectodermal cells/neurons (βIII tubulin), mesodermal cells/cardiac cells (cTroponin T), and endodermal cells (Pdx1; Fig. 3A). Then the miPSCs-O were microinjected into mouse blastocysts followed by transplanting into pseudopregnant mice. In 13-day post coitum (dpc) embryos, the iPSCs contributed to germline in the genital ridge as revealed by Oct4-EGFP expression (Fig. 3B, 3B′). We obtained three adult chimaeras from the miPSC-O as determined by coat color (Fig. 3C). Thus, the miPSC-O possesses differentiation potentials similar to mESCs.
Adult mice have been derived entirely from four factor-induced iPSCs via tetraploid complementation assays [25, 26]. It remained a significant question whether iPSCs generated by fewer reprogramming factors (e.g., Oct4 single factor), especially facilitated by small molecules, could pass this most stringent test of pluripotency. To address this, we further evaluated the pluripotency of our miPSC-O cells through tetraploid complementation. The miPSC-O cells were injected into CD-1 tetraploid embryos. We obtained one liveborn pup (iPSC 4N-comp pup) from 135 injected embryos for clone 1 (0.7%) and 10 pups from 125 injected embryos for clone 2 (8.0%; Fig. 3D), which is comparable with the efficiency of the four factor-induced iPSCs [25, 26] (Table 1). Simple sequence length polymorphism analysis followed by PCR of various marker genes was performed to verify the lineage of the miPSC-O and iPSC 4N-comp pups. The iPSC 4N-comp pups had the same profiles as the miPSC-O lines from which they were derived, but differed from the CD-1 donor of tetraploid embryos (Fig. 3E), demonstrating that the pups were entirely derived from the miPSC-O cells but not from the CD-1 donor embryos.
Table 1. Summary of embryo injections for tetraploid complementation
—, data not available.
Ten iPSC 4N-comp pups died at postnatal day 1 (P1) or P2. One pup did survive till P4 but was eaten by the foster mother. Although we have not obtained adult mice from the tetraploid complementation, the results demonstrated that the embryos derived from the miPSC-O cells were able to develop till delivery and gave rise to liveborn mice.
Our studies demonstrated that Oct4 in conjunction with A-83-01/AMI-5 treatment is sufficient to generate authentic iPSCs from MEFs. The miPSCs-O expressed typical pluripotency markers, differentiated into cell types of the three germ layers, produced live chimeras, and contributed to germline as mESCs. Until this study, iPSCs generated with fewer reprogramming factors (and treated with reprogramming enhancing small molecules) had not been tested with the most stringent pluripotency criteria, that is, tetraploid complementation, which represents a road block for future applications of small molecule-enabled reprogramming methods. We demonstrated that the miPSC-O cells could generate all-iPSC pups although they did not survive longer than P4. Genetic background of the MEFs from which the iPSCs are derived, or mutations acquired in long-term culture while reprogramming, were shown to affect the growth and behaviors of mice derived from four factor-induced iPSCs. They could affect the survival of iPSC pups as well. Alternatively, more iPSC clones may be tested to verify the full developmental ability of Oct4-induced iPSCs.
We for the first time demonstrated that AMI-5 facilitated reprogramming in the single factor Oct4 context. Since the report of generation of iPSCs by Oct4 and Klf4 with small molecules more than two years ago, it had been a significant challenge to further reduce the number of exogenous reprogramming genes from two to one. Concurrent with our work using only two small molecules, combination of four small molecules was shown to enable Oct4-induced reprogramming of MEFs . Given that TGF-β receptor inhibitors were used in both conditions for Oct4-induced reprogramming, it seems that AMI-5 has a remarkable reprogramming enhancing activity comparable with three-compound (Parnate, VPA, and CHIR99021) combination. AMI-5 inhibits the activities of PRMT 1/3/4/6, which belong to a family of proteins that catalyze the monomethylation or dimethylation of the arginine side chains. This post-translational modification is implicated in RNA processing, DNA repair, signal transduction, protein translocation, chromatin remodeling, and transcriptional regulations [27–29]. Further experiments are required to elucidate how AMI-5 facilitates the Oct4-induced reprogramming.
We thank Drs. Hongyan Zou, Janghwan Kim, Tongxiang Lin, Tianhua Ma, Wenlin Li, and other Ding Laboratoty members for suggestions and discussions. S.D. is supported by funding from NICHD, NHLBI, and NIMH/NIH, California Institute for Regenerative Medicine, Prostate Cancer Foundation, Fate Therapeutics, Esther B. O'Keeffe Foundation, and The Scripps Research Institute. This study was also supported by grants from National Science Foundation of China 90919060 and China National Basic Research Program 2011CB965301.
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
S.D., X.Y., and S.Z. are investigators of the Scripps Research Institute. Q.Z., X.Z., and H.W. are investigators of State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences. S.D. is a cofounder of Fate Therapeutics Inc. and Stemgent Inc.